The long-term goal of this research-training proposal is to understand how transcriptional feedback circuits regulate phenotype in mammalian systems. The specific goal is to understand how the transcriptional circuitry of human cytomegalovirus (CMV), a herpesvirus and important human pathogen, controls the virus's diverse replication phenotypes (CMV rapidly lyses fibroblasts, persistently infects endothelial cells, slowly replicates in epithelial cells, and enters latency in CD34+ progenitor cells). The mechanisms driving these diverse viral phenotypes (especially latency) remain unknown. However, the first viral circuit expressed upon infection and latent reactivation is CMV's auto-regulatory "master circuit", a crossed positive- and negative-feedback circuit called the Major Immediate-Early (MIE) circuit. The MIE circuit is known to initiate the viral transcriptional cascade and drive subsequent viral replication. I hypothesize that the MIE circuit might also control CMV's diverse replication phenotypes, including latency. To understand how MIE feedback might function as a genetic switch regulating entry into and exit from different replicative phenotypes, I will utilize a coupled experimental &theoretical approach to quantitatively characterize MIE circuit components, dynamics, feedback strength, cooperativity, and circuit noise structure in diverse cell types. If the MIE circuit controls CMV's diverse replication phenotypes, these experimental measurements, and the parallel mathematical modeling studies, are expected to yield different MIEfeedback dynamics in diverse cell types (e.g. pulses vs. stable oscillations vs. fixed states). The specific aims are: (1) to characterize the feedback dynamics and function of isolated MIE circuit components in diverse cell-types (i.e. outside the context of viral infection);(2) to map MIE feedback architecture &noise structure to replication phenotype in the full, intact MIE circuit during wild-type CMV infection (i.e.inside the context of viral infection);and (3) to elucidate MIE feedback kinetics in-vivo during infection of a murine model. This project will capitalize on recent advancesin single-cell microscopy, genetically encoded fluorescent probes, and automated image-analyses to quantify MIEfeedback kinetics in live single cells. I will gain training in whole-animaland single-cell in-vivo imaging techniques, and recently developed gene expression noise frequencyanalysesfor probing transcriptional feedback architecture in single-cells. Finally, I will also be trained in viral molecular biochemistry, viral mutagenesis, and viral "recombineering" techniques. This project should drive a new generation of theoretical models by providing a new conceptual model for mammalian transcriptional auto-regulationand a new experimental system for quantitativelyanalyzinghow feedback circuitry drives mammalian phenotype.